Lay Summary: I review the history of Oncotype and comment on the current status of this no longer controversial test.
The published literature for Oncotype primarily consists of validation studies to identify the optimal set of cancer-related genes. For example, tumor tissue can be retrieved from stored paraffin blocks of resected breast cancer, analyzed for patterns of genetic expression, which are then correlated with the patients’ outcomes and recorded in patient registries. In an initial study, van de Vijver and colleagues developed a classification system of genetic expression that outperformed all clinical variables in predicting the likelihood of developing distant metastases within 5 years. This genetic “signature” was then assayed in 295 patients with Stage I or II breast cancer; 151 had lymph node negative disease and 144 had positive nodes. The median duration of follow-up was 7.8 years for those who did not develop metastases. Patients assigned to a low-risk group by gene expression had a higher likelihood of metastasis-free survival than those classified by conventional criteria. Similarly, patients assigned to high-risk groups by gene expression had a higher risk of distant metastases than those classified conventionally. The authors concluded that the gene expression profile was a more powerful predictor of the outcome of disease than standard systems using clinical and histological criteria.
Two abstracts of validation studies were presented at the 2003 annual meeting of the American Society of Clinical Oncology. Expression of different genes was associated with either shorter or longer disease-free survival. Much of the work has been done by Paik. Paik and colleagues reported on a further validation study in 242 patients with node-negative breast cancer. Patients were identified from the databases and stored pathology specimens from prior NSABP clinical trials focusing on patients with estrogen receptor-positive and node-negative tumors. One of the purposes of the study was to identify a single multi-gene assay that could be subjected to formal clinical validation. Analysis indicated that of the 185 genes tested, 41 were associated with relapse-free survival. Many of the genes were closely correlated, and ultimately the authors identified a single multi-gene model containing 21 genes. This 21-gene predictor for likelihood of disease was prospectively studied in a formal clinical validation study of over 500 patients, the results of which were recently published. The difference in the risk of distant recurrence between patients with low recurrence scores and those with high recurrence scores was large and statistically significant. Fifty-one percent of patients were categorized as having a low risk, and their rate of distant recurrence at 10 years was 6.8 percent; 27 percent were categorized as having a high risk, with a rate of distant recurrence of 30.5 percent at 10 years. While this study reports important data, it remains unclear how this information can be used to either select or deselect patients for adjuvant therapy after primary surgical excision of breast cancer. In fact, the authors state clearly in their summary, “The current data cannot be used to select women for tamoxifen therapy.” Further studies are needed, focusing on how test results may be used in the clinical management of the patient.
In February 2005, the BlueCross BlueShield Association Technology Evaluation Center (TEC) conducted a technology assessment on gene expression profiling for managing breast cancer treatment. The TEC Assessment summarized the evidence for four different gene expression profiling assays, in various stages of development, that are intended for eventual use in identifying those patients at low risk of recurrence for whom adjuvant chemotherapy can be avoided. The four assays were the 21-gene Oncotype DX (Genomic Health), the 70-gene MammaPrint® (Agendia; also referred to as the “Amsterdam signature”), the 76-gene “Rotterdam signature” (Veridex), and a 41-gene signature reported by Ahr and colleagues. Because current selection methods by traditional clinical and histopathological criteria exclude only a small proportion from adjuvant chemotherapy, and only a small proportion derive significant benefit, it is likely that a significant number of women could avoid the side effects of chemotherapy if more accurate selection methods were available. However, the TEC Assessment concluded that because published evidence supporting clinical utility is not available, the evidence for all of the gene expression panels is insufficient to permit conclusions concerning the effect of gene expression profiling on selecting patients who do not need chemotherapy for the purpose of avoiding adverse outcomes, while maintaining or improving disease-free or overall survival outcomes.
The TEC Assessment further evaluated several studies of gene expression profiling of pretreatment tumor samples to predict response to adjuvant or neoadjuvant chemotherapy regimens, including the December 2004 publication by Paik and colleagues discussed above. Most studies were first reports of the development of the gene expression signatures from sample test sets. Some included small, within-institution studies of separate validation specimens. Outcomes reported were intermediate measures of response to chemotherapy. No studies linked the results of gene expression signatures to disease-free or overall survival outcomes, thus the evidence was insufficient to permit conclusions regarding the use of gene expression profiling to improve the selection of beneficial chemotherapy regimens and improve disease-free or overall survival outcomes.
In 2006, the National Comprehensive Cancer Network published the following statement in their breast cancer guidelines: “Limited data supports that a 21 gene RT-PCR assay (Oncotype Dx) may provide both prognostic information and prediction of benefit, or lack thereof, from chemotherapy in women with axillary lymph node negative, hormone receptor-positive breast cancer treated with tamoxifen. The panel awaits additional studies prior to issuing a specific recommendation regarding this or similar assays.”
An updated search of the MEDLINE database failed to identify any new peer-reviewed publications which alter the conclusions reached above. Published abstracts and meeting presentations continue to focus on retrospective analyses of banked tumor tissue. However, I did find that BC Medical Advisory Board reversed itself in June 2007 and now considers Oncotype to be medically necessary.
Recently started is the first prospective, randomized clinical trial of Oncotype Dx™. The protocol, under the National Cancer Institute’s (NCI) Program for the Assessment of Clinical Cancer Tests, is being reviewed by the NCI Central Institutional Review Board. Named TAILORx (Trial Assigning Individualized Options for Treatment), this multi-institutional trial will focus on patients with ER-positive, node negative tumors. Patients will be categorized using Oncotype DX results and new risk definitions: low recurrence score (RS) <11; intermediate RS 11-25; and high risk RS >25. Thus, more patients will be assigned to intermediate risk, the only category that will be randomized to receive hormonal therapy with or without adjuvant therapy. Those patients assigned to high risk status will receive hormonal therapy and adjuvant chemotherapy, while those at lowest risk will receive only hormonal therapy. Plans are to enroll between 9,000 and 20,000 women and follow them for cancer recurrence.
The scientific evidence is mixed. There is no phase III evidence (TAILORX is ongoing) and there is validation and retrospective reviews. The American Society of Clinical Oncology (ASCO) has included Oncotype DX™ into its 2007 updated clinical guidelines for the use of tumor markers in breast cancer. Most recently, NCCN has decided to include Oncotype in its 2008 guidelinesand BCBS issued a supportive TEC assessment. NCCN sets the tumor size at 0.5cm.
There are many insurers who decided to cover but even where there is not specific policy to cover, with the NCCN decision, I consider the evidence to show that Oncotype testing is no longer investigational.
I. Baker J et al. Tumor gene expression predicts distant disease-free survival (DDFS) in breast cancer patients with 10 or more positive nodes: High throughput RT-PCR assay of paraffin-embedded tumor tissues. Proc Am Soc Clin Oncol 2003;22:850 (abstract 3415)
Paik S, Shak S, Tang G et al. Multi-gene RT-PCR assay for predicting recurrence in node negative breast cancer patients – National Surgical Adjuvant Breast and Bowel Project (NSABP) studies B-20 and B-14. San Antonio Breast Cancer Symposium. San Antonio, TX, December 2003.
Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. NEJM 2004;351(27):2817-26
2007 TEC Assessment: Gene expression profiling for managing breast cancer treatment. www.bcbs.com/tec/vol20/20_03.html
National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology. Breast Cancer. http://nccn.org/professionals/physician_gls/PDF/breast.pdf, BINV-6.
Eileen Rakovitch et al, A population-based validation study of the DCIS Score predicting recurrence risk in individuals treated by breast-conserving surgery alone. Breast Cancer Research and Treatment July 2015, Volume 152, Issue 2, pp 389-39